Dr. Gary Heller
April 12, 2011
It seems as if Dr. Gary Heller was born to do radio research of some kind. Boasting a Ph.D in Communication from USC - with a dissertation that dealt specifically with the emotional reactions that people have for music and radio - Heller hooked up with CBS Radio soon thereafter, where he's been analyzing data from Arbitron, the PPM, various music formats, Qualitative and Sales. Here Heller explains his work and how he balances his analysis with successful radio programming instincts.
How has your responsibilities grown or changed since you first started doing research at CBS?
My research responsibilities have grown substantially since I started with CBS Radio 10 years ago. I started out on the Sales Research side at KNX 1070 Newsradio in Los Angeles under Mike Masterson and George Nicholaw. However, because of my programming background and interest I was always helping out the KNX programmers as well as all the sales people.
In 2004, there was an opening for the Format VP in charge of Research & Audience Measurement, and Pres./Programming Steve Rivers gave me those Programming Research responsibilities. In 2005, newly-hired Head of Research Tony Jarvis added National Research responsibilities to my plate. When he left, I inherited his Media Rating Council and other duties.
A couple years later when Lucy Hughes came over to head CBS Radio Research, my KNX duties were dropped and I became 100% Corporate. Since then, it continues to evolve, adapting to whatever is needed (like the introduction of PPM). Most recently I've added Digital responsibilities. Today, I am blessed to be able to work with great radio people like Dan Mason, Scott Herman, Greg Strassell, Kevin Weatherly, Chris Oliviero, Lucy Hughes and all of our Market Managers and Program Directors.
If it means helping out the ratings or making money for the company, I help out in any way I can.
Let's get to the introduction of PPM. Do you feel your research has, by now, developed a firm grip on what makes PPM "tick" in terms of programming strategies that would optimize ratings, or are there still areas to figure out?
All the PPM tactics in the world won't help unless you have great programmers steering the stations. The research that's been provided helps them navigate those waters, but it's just one of the elements that they take into consideration. As we evolve deeper into PPM, there is more and more information, and they will continue to need someone to distill it down and give it meaning. I am sure that will continue into the future. I am very proud of how CBS Radio has transitioned under PPM.
A recent Coleman Research report, showcased during a recent Arbitron PPM client briefing, cautioned that programmers shouldn't make knee-jerk decisions on the basis of short-term PPM data. Do you agree ... and how do you recommend programmers analyze and use the research before them?
Good solid strategy that super-serves the listener should always guide a radio station. Knee-jerk reactions show a lack of faith in that strategy.
There can be a lot of noise in PPM data, and simple things like breaking down demos, for instance, only tend to increase that noise, making it harder to find truth. In addition, it also can be easy for a programmer to read too much into the data. So, when something is not clear, I encourage programmers to have someone like me take an objective impartial look at the numbers to see how they read.
I don't want to give away your trade secrets, of course, but how does anyone -- a programmer, a OM, GM, whomever -- learn how to separate "objective impartiality" from subjectivity in analyzing research?
Mostly from experience. It's a lot like Malcolm Gladwell talked about in his book "Outliers;" the more you do something, the better you'll get at it, and the really good people do it a lot. It helps to have an analytical mind that enjoys diving into the data; that helps the desire to experience.
Regarding digital: From your research, are those who listen to radio stations online and who interact with the station through Facebook and/or Twitter any different in terms of demography, listening habits and likes/dislikes? If so, what are the differences ... and what have CBS stations done to optimize that interaction?
It's a similar question to whether the small sub-section of listeners who called in to the station to request a song are truly representative of the listeners to the radio station itself. In some digital cases, they might be more similar to the smaller sub-section because of the uses and gratifications of the digital medium being used; in others, the tastes might be in line with the overall. It really depends upon the use. Regardless, if the demographic distribution of the digital medium skews differently than the distribution of the radio station, there is bound to be some taste or habit difference. I'd sure love to see the results of that research when you find it.
How does CBS Radio delineate its digital strategy?
As CBS Radio strives to super-serve its listeners on-air, we also aim to super-serve them in the digital realm. We have two organizations that seek to provide the best content online: CBSLocal and the CBS Interactive Music Group. CBSLocal teams up the assets of CBS' spoken word radio stations -- such as its News, News/Talk and Sports stations --with the local CBS television stations.
For instance, losangeles.cbslocal.com marries Channels 2 and 9 with KNX 1070 Newsradio and KFWB News/Talk 980 into the same platform to become the Los Angeles online source for news, information, entertainment and everything local. By combining all their resources, CBSLocal can provide the best content for our users.
Likewise, the CBS Radio Music stations are part of the CBS Interactive Music Group along with last.fm and mp3.com, radio.com, tonefuse, AOL and Yahoo. One of the things CBS IMG does to help create the best content possible is to aid each of the local music radio station sites by creating topical relevant web-friendly content that the stations can choose to use to make their own sites more dynamic, interesting and relevant.
CBS IMG adds special-event programming to its sites like the Live On Letterman series where major artists will play live and that will be broadcast on-air as well as available online. On April 12th, the Foo Fighters was part of the series. Plus, for those who want to hear audio customized to their own tastes, we have last.fm, which also has some of the best music info available anywhere. You can even listen to it on your Xbox if you wanted to. It's all about creating the best content.
What exactly is the "Alpha Boomer," as compared to a Baby Boomer, and what makes this consumer so sought-after?
Alpha Boomers are the first wave of the Baby Boomers, typically considered as 55-64 year olds...
One of the key things I'm interested in is making sure that the Alpha Boomer gets valued by advertisers properly. Radio does a great job of delivering this demo, yet advertisers don't value them since they are out of the 25-54 demo. That needs to change. That age cell is an active consumer (over-indexing in things like travel and luxury cars), with a larger household income than most cells, and they are not brand-loyal (they are willing to try new products). Radio can't afford to deliver quality audience and not get paid for them.
I'd like to see them incorporated into a new standard demo -- 25-64 -- which when you think about it really is the employed adult world ... and we know how well radio does with delivering employed people to advertisers. Oddly enough, it's been television that has been leading the way in pushing the Alpha Boomer as they try to get value beyond their 18-49 money demo. The population profile of the U.S. has changed ... and advertisers need to recognize that.
Another demo that had been ignored by radio until the advent of PPM is 6-11.How worthwhile is conducting research on that demo and altering your programming to cater to it?
Even though CBS Radio does have one Family Hits radio station -- KYDZ-A/Las Vegas, which does benefit from the 6-11 inclusion -- pre-teen is not the first or second demo I look at to judge how well our stations are doing. It's nice to see but certainly not surprising to find out that 6-11 year-olds like hit radio, and I will make sure that their cume is included when I quote our Top 40s' total cume. Beyond that, it doesn't have the greatest utility for me.
There has been some debate over whether having original programming, be it niche formats on HD side channels, or original content on digital platforms, can drain the audience away from terrestrial. Has your research confirmed or debunked that?
While we know that there will be greater duplication between radio and the new media, we also know that having programming available on different platforms increases the overall use of the media and increases the connection with the brand, which then builds the goodwill of the brand, aiding it being top-of-mind, and recycles listening back through to the station. So greater duplication does not necessarily mean cannibalization.
One last question: Speaking as someone who's spent years analyzing research, exactly where does "gut" fit into the mix? Through your ongoing discussions with programmers, have you reached a consensus on how to use "gut instinct" with research, be it a counterweight or just as something impulsive? Or is the whole notion of using your "gut" more of an Urban legend these days?
Programming is a mixture of art and science ... and if only one is used, you're doomed, so I absolutely emphasize to program directors to trust their gut. I think it's imperative to involve it in the process. When looking at PPM data, if the gut and the data agree, you can feel pretty safe. However, if the gut and the data disagree, I lean towards the gut as there can be lots of noise with PPM data.
Gut is useful in other ways as well: Gut directs one to mine for things in data ... gut is the sniff test that the data is alright when one receives research from the field ... and gut tends to be more all-encompassing, feeling through all the ramifications of the decision. Programmers have trained their guts over many years, and that tool should not be dismissed.